Atrial fibrillation, the most common heart-rhythm disorder, increases the risk of ischemic stroke by 3- to 5-fold. Anticoagulant therapy has been proven to prevent 60-80% of ischemic strokes that would otherwise occur from atrial fibrillation. Patients with a history of intracerebral hemorrhage have been excluded from clinical trials of anticoagulation in patients with atrial fibrillation. Whether to use anticoagulation in these patients represents a major knowledge gap and clinical dilemma. Many clinicians fear that the proven benefit of anticoagulation in preventing ischemic stroke will be offset by an increase in hemorrhagic stroke. Preliminary data from multicenter studies indicate that, in patients with atrial fibrillation and recent intracerebral hemorrhage, anticoagulation is associated with a decreased risk of ischemic stroke and overall mortality with no offsetting increase in the burden of recurrent hemorrhagic stroke. These pilot data also suggest that anticoagulation is associated with better long-term functional outcomes. Although these data are compelling, they are observational findings and subject to confounding and bias. In clinical practice, about one-third of patients with intracerebral hemorrhage and atrial fibrillation receive anticoagulation and the remainder receive antiplatelet therapy such as aspirin. Current clinical guidelines call for randomized, blinded clinical trials to provide high-quality evidence to determine the best treatment. The argument for a clinical trial is made more compelling by the advent of apixaban, a relatively new oral anticoagulant drug, which was found to have similar bleeding risks as aspirin in a recent head-to-head trial. This application is for a multicenter, double-blinded, randomized clinical trial of apixaban versus aspirin in patients with atrial fibrillation and a recent intracerebral hemorrhage. Seven hundred patients will be recruited and followed for 1 to 3 years at 125 sites in NINDS StrokeNet. The aim of the study will be to test the hypothesis that apixaban improves outcomes compared to aspirin. The primary outcome will be any stroke (ischemic or hemorrhagic) or death. This composite outcome addresses both efficacy and safety, and represents a clinically meaningful endpoint often used in stroke prevention trials. The secondary endpoint will be functional status, as measured by the modified Rankin Scale score. This secondary endpoint will be used to test the hypothesis that apixaban not only reduces stroke recurrence but also the severity of any strokes that do occur. If confirmed, this novel finding would provide an important patient-centered context for the primary trial results. This proposal offers an opportunity to improve the outcomes of patients with intracerebral hemorrhage by focusing on their brain health both in the acute setting and over the long term. The impact of this study is that its successful completion, regardless of the direction of its results, would immediately guide clinical care and set the stage for future trials of antithrombotic therapy in patients with atrial fibrillation and other types of bleeding.